Pregnant While on a GLP-1 Microdose? What the Birth Defect Studies and the Label Say About Early Exposure

Getting pregnant while on a GLP-1 microdose happens more often than people expect. Weight loss can make cycles more regular, some people assume a low dose cannot matter, and tirzepatide can make the pill less reliable. Early GLP-1 exposure in pregnancy is now common enough that researchers have studied it in several hundred thousand pregnancies.

This article covers what those studies found about birth defects, pregnancy loss and fetal growth, what the FDA labels say, and why stopping the drug comes with its own trade-off. One limit applies throughout: every study used prescribed doses. None of them looked at sub-label microdoses, and none of this replaces a conversation with your own clinician.

How common is getting pregnant while on a GLP-1?

Pregnancy exposure to weight-loss medicines is still uncommon, but it nearly doubled in a single year. Haynie and colleagues (European Journal of Obstetrics, Gynecology, and Reproductive Biology, 2026) looked at about 2 million pregnancies in US commercial claims. Overall, 0.32% were exposed to an FDA-approved anti-obesity medicine before or during pregnancy. Exposure rose from 0.31% in 2022 to 0.58% in 2023, and in 2023 semaglutide overtook phentermine as the most common drug.

The same study found exposure was highest in women over 40 (1.1%) and in those with pre-pregnancy obesity (1.4%). A 2026 review in BJOG by Lau and colleagues adds that the true number is probably higher, because maternity services do not routinely ask about these drugs. Microdosers sourcing from compounding pharmacies may be even less visible in claims data.

Do GLP-1 medicines in early pregnancy cause birth defects?

The best large study found no clear increase in major birth defects once maternal diabetes was taken into account. Cesta and colleagues (JAMA Internal Medicine, 2024) pooled health records from 4 Nordic countries, the US MarketScan database and Israel's Maccabi system. They counted a pregnancy as exposed if a prescription was filled any time from 90 days before conception to the end of the first trimester.

At first glance the chart looks worrying. Major malformations were found in 8.3% of the 938 GLP-1 exposed infants, against 3.7% of all infants. But the right comparison is not the general population. Infants of mothers with type 2 diabetes had a 5.3% rate overall, and insulin-exposed infants, whose mothers had diabetes serious enough to need it, had 7.8%. Diabetes itself raises the risk of birth defects, so the researchers compared each drug with insulin.

Against insulin, the adjusted relative risk for GLP-1 receptor agonists was 0.95, with a confidence interval of 0.72 to 1.26. The authors concluded the results did not indicate a large increased risk beyond what diabetes itself confers, while noting that some estimates were imprecise and that confirmation was needed. For people microdosing without diabetes, this study is not a perfect match, which is why the newer research below matters.

Is it worse to keep taking a GLP-1 into the first trimester?

Continuing a GLP-1 after conception did not clearly raise risk compared with stopping beforehand, but the data cannot exclude a real difference. Brown and colleagues (Annals of Internal Medicine, 2026) used US claims from 2011 to 2024 to emulate a trial. They took 3572 pregnancies in which a GLP-1 was dispensed in the 90 days before the last menstrual period, and compared those who kept filling prescriptions into the first trimester with those who did not. About 41.1% of the women had type 2 diabetes.

The weighted risk of a pregnancy not ending in a live birth was 29.7% with continuation and 27.1% without, an adjusted risk ratio of 1.09 (95% CI 0.98 to 1.23). Among the live births linked to an infant record, the ratios for continuation versus stopping were all above 1, but every interval crossed it.

The authors were careful with their wording. The estimates for malformations and small-for-gestational-age babies, they wrote, were "compatible with both no increased risk and clinically relevant differences in risk." In plain terms, this is not an all-clear. It is an absence of a clear signal, from a study that was not large enough to detect a modest one.

What do the 2026 meta-analyses say about GLP-1 exposure in pregnancy?

Two 2026 meta-analyses reached the same cautious answer: no consistent link with major birth defects or pregnancy complications. Liu and colleagues (Endocrine Connections, 2026) pooled 6 studies with 286,599 women, 43,577 of them exposed within three months before conception or during pregnancy. The relative risk of major malformations was 1.02 (95% CI 0.96 to 1.08).

Large-for-gestational-age babies were left off the chart because the estimate was so uncertain. The relative risk was 2.31, but the interval ran from 0.56 to 9.44, which means the data are consistent with anything from a lower risk to a much higher one. Liu and colleagues closed by saying the findings "should not be interpreted as proof of safety."

Hakim and colleagues (American Journal of Obstetrics and Gynecology, 2026) cast a wider net: 22 studies covering 49,395 exposed pregnancies, with 10 pooled in a meta-analysis. They found no significant association with major or cardiac malformations, preterm delivery, pregnancy loss or high blood pressure in pregnancy. One signal did appear, for kidney malformations (odds ratio 1.23, 95% CI 1.09 to 1.39). It was largely driven by one large cohort with big baseline differences between groups, and the authors think it likely reflects maternal disease rather than the drug. Their summary: cautious reassurance after inadvertent exposure, but no support for routine use in pregnancy.

Why does the label say to stop semaglutide 2 months before pregnancy?

The 2-month rule exists because semaglutide leaves the body slowly. The Wegovy prescribing information gives an elimination half-life of approximately 1 week and says semaglutide is present in the circulation for about 5 to 7 weeks after the last 2.4 mg injection. It tells clinicians to discontinue Wegovy at least 2 months before a planned pregnancy to account for that.

The warning itself rests on animal studies. In rabbits and monkeys given semaglutide during organ development, the label reports early pregnancy losses and structural abnormalities, which coincided with marked weight loss in the mothers. The label also states plainly that "Weight loss offers no benefit to a pregnant patient and may cause fetal harm," and that human data are insufficient to establish a drug-associated risk.

Tirzepatide clears a little faster. The Zepbound prescribing information gives a half-life of approximately 5 days and advises stopping when a pregnancy is recognized. A smaller weekly dose does not change the half-life, so a microdose still lingers for weeks, just at lower levels. How much that lower level matters for a pregnancy has never been studied.

For context, the Wegovy label puts the background risk in the US general population at 2% to 4% for major birth defects and 15% to 20% for miscarriage in clinically recognized pregnancies. Those risks exist with or without any medicine. Our guide to GLP-1 microdosing before IVF covers planned conception and washout timing in more depth.

Does stopping a GLP-1 in pregnancy carry its own risks?

Possibly, and it is the part of the story most people miss. Yu and colleagues (Obstetrics and Gynecology, 2026) used linked health records to compare 429 women exposed to semaglutide into pregnancy, 801 former users who stopped before pregnancy, and 2,203 nonusers, all with overweight or obesity. The exposed group had a median of 44 days of exposure during pregnancy.

Both semaglutide groups had higher odds of excessive gestational weight gain, gestational diabetes, excess fetal growth and cesarean delivery than nonusers. The striking finding is that women exposed into pregnancy and women who had stopped before it looked the same. The authors concluded the extra risk is "more likely attributable to rebound effects after treatment discontinuation rather than to intrauterine exposure itself."

That does not mean anyone should keep taking a GLP-1 while pregnant. It means the weeks after stopping deserve a plan: early glucose screening if your clinician advises it, guidance on healthy weight gain, and support for the return of appetite. Our article on when the hunger comes back explains why appetite rebounds after a GLP-1 is stopped.

What should you do if you find out you are pregnant on a GLP-1 microdose?

Stop the next dose, call your prescriber or an obstetric clinician, and avoid assuming the worst. Both labels advise discontinuing once a pregnancy is recognized. The BJOG review by Lau and colleagues recommends individualized counselling "rather than assuming that exposure is either established to be harmful or proven to be safe."

Steps that follow from the evidence:

• •Write down the drug, the dose, whether it was compounded, and the dates of your last few injections. Clinicians use this to estimate when exposure ended.
• •Ask about the pregnancy exposure registry for your drug. Novo Nordisk and Eli Lilly both list one in their prescribing information.
• •Ask whether earlier screening for gestational diabetes makes sense, given the rebound findings.
• •If you were on tirzepatide and the pill, tell the clinician. The Zepbound label advises a non-oral or backup method for 4 weeks after starting and after each dose increase. Our birth control guide covers this.
• •Plan ahead for after delivery too. Our breastfeeding article reviews what is known about restarting while nursing.

Frequently asked questions

Clinical references

1. Cesta CE, Rotem R, Bateman BT, et al. Safety of GLP-1 Receptor Agonists and Other Second-Line Antidiabetics in Early Pregnancy. JAMA Internal Medicine. 2024. doi:10.1001/jamainternmed.2023.6663
2. Brown JP, Huybrechts KF, Straub L, et al. Continuing Glucagon-Like Peptide-1 Receptor Agonists Into the First Trimester of Pregnancy and Pregnancy Outcomes: A Target Trial Emulation Study Using Claims Information. Annals of Internal Medicine. 2026. doi:10.7326/ANNALS-25-04820
3. Liu X, Xiong B, Yang R, et al. Periconceptional use of GLP-1 receptor agonists and the risk of major congenital malformations: a systematic review and meta-analysis. Endocrine Connections. 2026. doi:10.1530/EC-26-0280
4. Hakim J, Rajesh D, Tello JA. Neonatal and obstetric outcomes following periconceptional exposure to glucagon-like peptide-1 receptor agonists: a systematic review and meta-analysis. American Journal of Obstetrics and Gynecology. 2026. doi:10.1016/j.ajog.2026.04.037
5. Yu Y, Li X, Groth SW, et al. Gestational Weight Gain and Pregnancy Outcomes After Semaglutide Exposure. Obstetrics and Gynecology. 2026. doi:10.1097/AOG.0000000000006325
6. Haynie C, Musty A, Choi H, et al. Prevalence and temporal trends of exposure to anti-obesity medications during preconception and pregnancy in a nationwide US claims database. European Journal of Obstetrics, Gynecology, and Reproductive Biology. 2026. doi:10.1016/j.ejogrb.2026.115503
7. Lau KGY, Chopra A, Mullins E, et al. Glucagon-Like Peptide-1 Receptor Agonists in Pregnancy: Periconception Exposure and Perinatal Outcomes. BJOG. 2026. doi:10.1111/1471-0528.70339
8. WEGOVY (semaglutide) prescribing information, sections 8.1 and 8.3. DailyMed. dailymed.nlm.nih.gov
9. ZEPBOUND (tirzepatide) prescribing information, section 8.1. DailyMed. dailymed.nlm.nih.gov